skills/utkusen/sast-skills/sast-xss/Gen Agent Trust Hub

sast-xss

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8). It ingests untrusted data from the analyzed codebase and sast/architecture.md and interpolates this content into instructions for subagents without the use of boundary markers or explicit safety warnings. This could allow an attacker to embed instructions in source code comments that mislead the agent, such as instructing it to skip certain files or report no findings. \n
  • Ingestion points: Project source code and sast/architecture.md are read and passed to subagents as context. \n
  • Boundary markers: Absent. The instructions do not define clear delimiters or "ignore embedded instructions" warnings when passing external data to subagents. \n
  • Capability inventory: The skill performs file reads on the codebase and writes results into the sast/ directory. \n
  • Sanitization: Absent. No escaping or filtering of external content is specified before interpolation into agent prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 08:00 AM